Handling missing quality of life data in HIV clinical trials

What is practical?

Diane L. Fairclough, Herbert Thijs, I. Chan Huang, Henrik W. Finnern, Albert W Wu

Research output: Contribution to journalArticle

Abstract

Aims: Missing health-related quality of life (HRQOL) data in clinical trials can impact conclusions but the effect has not been thoroughly studied in HIV clinical trials. Despite repeated recommendations to avoid complete case (CC) analysis and last observation carried forward (LOCF), these approaches are commonly used to handle missing data. The goal of this investigation is to describe the use of different analytic methods under assumptions of missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR) using HIV as an empirical example. Methods: Medical Outcomes Study HIV (MOS-HIV) Health Survey data were combined from two large open-label multinational HIV clinical trials comparing treatments A and B over 48 weeks. Inclusion in the HRQOL analysis required completion of the MOS-HIV at baseline and at least one follow-up visit (weeks 8, 16, 24, 40, 48). Primary outcomes for the analysis were change from week 0 to 48 in mental health summary (MHS), physical health summary (PHS), pain and health distress scores analyzed using CC, LOCF, generalized estimating equations (GEE), direct likelihood and sensitivity analyses using joint mixed-effects model, and Markov chain Monte Carlo (MCMC) multiple imputation. Time and treatment were included in all models. Baseline and longitudinal variables (adverse event and reason for discontinuation) were only used in the imputation model. Results: A total of 511 patients randomized to treatment A and 473 to treatment B completed the MOS-HIV at baseline and at least one follow-up visit. At week 48, 71% of patients on treatment A and 31% on treatment B completed the MOS-HIV survey. Examining changes within each treatment group, CC and MCMC generally produced the largest or most positive changes. The joint model was most conservative; direct likelihood and GEE produced intermediate results; LOCF showed no consistent trend. There was greater spread for within-group changes than between-group differences (within MHS scores for treatment A: -0.1 to 1.6, treatment B: 0.4 to 2.0; between groups: -0.7 to 0.4; within PHS scores for treatment A: -1.5 to 0.4, treatment B: -1.7 to -0.2; between groups: 0.1 to 1.1). The size of within-group changes and between-group differences was of similar magnitude for the pain and health distress scores. In all cases, the range of estimates was small

Original languageEnglish (US)
Pages (from-to)61-73
Number of pages13
JournalQuality of Life Research
Volume17
Issue number1
DOIs
StatePublished - Feb 2008

Fingerprint

Quality of Life
Clinical Trials
HIV
Outcome Assessment (Health Care)
Therapeutics
Markov Chains
Health
Observation
Mental Health
Joints
Pain
Health Surveys

Keywords

  • HIV clinical trials
  • Missing data
  • Quality of life

ASJC Scopus subject areas

  • Rehabilitation
  • Nursing(all)

Cite this

Handling missing quality of life data in HIV clinical trials : What is practical? / Fairclough, Diane L.; Thijs, Herbert; Huang, I. Chan; Finnern, Henrik W.; Wu, Albert W.

In: Quality of Life Research, Vol. 17, No. 1, 02.2008, p. 61-73.

Research output: Contribution to journalArticle

Fairclough, Diane L. ; Thijs, Herbert ; Huang, I. Chan ; Finnern, Henrik W. ; Wu, Albert W. / Handling missing quality of life data in HIV clinical trials : What is practical?. In: Quality of Life Research. 2008 ; Vol. 17, No. 1. pp. 61-73.
@article{6cdb61be4b994a2ca07b3e76b1300315,
title = "Handling missing quality of life data in HIV clinical trials: What is practical?",
abstract = "Aims: Missing health-related quality of life (HRQOL) data in clinical trials can impact conclusions but the effect has not been thoroughly studied in HIV clinical trials. Despite repeated recommendations to avoid complete case (CC) analysis and last observation carried forward (LOCF), these approaches are commonly used to handle missing data. The goal of this investigation is to describe the use of different analytic methods under assumptions of missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR) using HIV as an empirical example. Methods: Medical Outcomes Study HIV (MOS-HIV) Health Survey data were combined from two large open-label multinational HIV clinical trials comparing treatments A and B over 48 weeks. Inclusion in the HRQOL analysis required completion of the MOS-HIV at baseline and at least one follow-up visit (weeks 8, 16, 24, 40, 48). Primary outcomes for the analysis were change from week 0 to 48 in mental health summary (MHS), physical health summary (PHS), pain and health distress scores analyzed using CC, LOCF, generalized estimating equations (GEE), direct likelihood and sensitivity analyses using joint mixed-effects model, and Markov chain Monte Carlo (MCMC) multiple imputation. Time and treatment were included in all models. Baseline and longitudinal variables (adverse event and reason for discontinuation) were only used in the imputation model. Results: A total of 511 patients randomized to treatment A and 473 to treatment B completed the MOS-HIV at baseline and at least one follow-up visit. At week 48, 71{\%} of patients on treatment A and 31{\%} on treatment B completed the MOS-HIV survey. Examining changes within each treatment group, CC and MCMC generally produced the largest or most positive changes. The joint model was most conservative; direct likelihood and GEE produced intermediate results; LOCF showed no consistent trend. There was greater spread for within-group changes than between-group differences (within MHS scores for treatment A: -0.1 to 1.6, treatment B: 0.4 to 2.0; between groups: -0.7 to 0.4; within PHS scores for treatment A: -1.5 to 0.4, treatment B: -1.7 to -0.2; between groups: 0.1 to 1.1). The size of within-group changes and between-group differences was of similar magnitude for the pain and health distress scores. In all cases, the range of estimates was small",
keywords = "HIV clinical trials, Missing data, Quality of life",
author = "Fairclough, {Diane L.} and Herbert Thijs and Huang, {I. Chan} and Finnern, {Henrik W.} and Wu, {Albert W}",
year = "2008",
month = "2",
doi = "10.1007/s11136-007-9284-3",
language = "English (US)",
volume = "17",
pages = "61--73",
journal = "Quality of Life Research",
issn = "0962-9343",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Handling missing quality of life data in HIV clinical trials

T2 - What is practical?

AU - Fairclough, Diane L.

AU - Thijs, Herbert

AU - Huang, I. Chan

AU - Finnern, Henrik W.

AU - Wu, Albert W

PY - 2008/2

Y1 - 2008/2

N2 - Aims: Missing health-related quality of life (HRQOL) data in clinical trials can impact conclusions but the effect has not been thoroughly studied in HIV clinical trials. Despite repeated recommendations to avoid complete case (CC) analysis and last observation carried forward (LOCF), these approaches are commonly used to handle missing data. The goal of this investigation is to describe the use of different analytic methods under assumptions of missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR) using HIV as an empirical example. Methods: Medical Outcomes Study HIV (MOS-HIV) Health Survey data were combined from two large open-label multinational HIV clinical trials comparing treatments A and B over 48 weeks. Inclusion in the HRQOL analysis required completion of the MOS-HIV at baseline and at least one follow-up visit (weeks 8, 16, 24, 40, 48). Primary outcomes for the analysis were change from week 0 to 48 in mental health summary (MHS), physical health summary (PHS), pain and health distress scores analyzed using CC, LOCF, generalized estimating equations (GEE), direct likelihood and sensitivity analyses using joint mixed-effects model, and Markov chain Monte Carlo (MCMC) multiple imputation. Time and treatment were included in all models. Baseline and longitudinal variables (adverse event and reason for discontinuation) were only used in the imputation model. Results: A total of 511 patients randomized to treatment A and 473 to treatment B completed the MOS-HIV at baseline and at least one follow-up visit. At week 48, 71% of patients on treatment A and 31% on treatment B completed the MOS-HIV survey. Examining changes within each treatment group, CC and MCMC generally produced the largest or most positive changes. The joint model was most conservative; direct likelihood and GEE produced intermediate results; LOCF showed no consistent trend. There was greater spread for within-group changes than between-group differences (within MHS scores for treatment A: -0.1 to 1.6, treatment B: 0.4 to 2.0; between groups: -0.7 to 0.4; within PHS scores for treatment A: -1.5 to 0.4, treatment B: -1.7 to -0.2; between groups: 0.1 to 1.1). The size of within-group changes and between-group differences was of similar magnitude for the pain and health distress scores. In all cases, the range of estimates was small

AB - Aims: Missing health-related quality of life (HRQOL) data in clinical trials can impact conclusions but the effect has not been thoroughly studied in HIV clinical trials. Despite repeated recommendations to avoid complete case (CC) analysis and last observation carried forward (LOCF), these approaches are commonly used to handle missing data. The goal of this investigation is to describe the use of different analytic methods under assumptions of missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR) using HIV as an empirical example. Methods: Medical Outcomes Study HIV (MOS-HIV) Health Survey data were combined from two large open-label multinational HIV clinical trials comparing treatments A and B over 48 weeks. Inclusion in the HRQOL analysis required completion of the MOS-HIV at baseline and at least one follow-up visit (weeks 8, 16, 24, 40, 48). Primary outcomes for the analysis were change from week 0 to 48 in mental health summary (MHS), physical health summary (PHS), pain and health distress scores analyzed using CC, LOCF, generalized estimating equations (GEE), direct likelihood and sensitivity analyses using joint mixed-effects model, and Markov chain Monte Carlo (MCMC) multiple imputation. Time and treatment were included in all models. Baseline and longitudinal variables (adverse event and reason for discontinuation) were only used in the imputation model. Results: A total of 511 patients randomized to treatment A and 473 to treatment B completed the MOS-HIV at baseline and at least one follow-up visit. At week 48, 71% of patients on treatment A and 31% on treatment B completed the MOS-HIV survey. Examining changes within each treatment group, CC and MCMC generally produced the largest or most positive changes. The joint model was most conservative; direct likelihood and GEE produced intermediate results; LOCF showed no consistent trend. There was greater spread for within-group changes than between-group differences (within MHS scores for treatment A: -0.1 to 1.6, treatment B: 0.4 to 2.0; between groups: -0.7 to 0.4; within PHS scores for treatment A: -1.5 to 0.4, treatment B: -1.7 to -0.2; between groups: 0.1 to 1.1). The size of within-group changes and between-group differences was of similar magnitude for the pain and health distress scores. In all cases, the range of estimates was small

KW - HIV clinical trials

KW - Missing data

KW - Quality of life

UR - http://www.scopus.com/inward/record.url?scp=37849007983&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=37849007983&partnerID=8YFLogxK

U2 - 10.1007/s11136-007-9284-3

DO - 10.1007/s11136-007-9284-3

M3 - Article

VL - 17

SP - 61

EP - 73

JO - Quality of Life Research

JF - Quality of Life Research

SN - 0962-9343

IS - 1

ER -